The first objective of this project is to determine the prevalence of behavior problems in three year old children living in a demarcated, rural community. The sample is drawn from permanent residents of Martha's Vineyard Island, located seven miles off the Massachusetts mainland. The population of the island is 8500, and about 100 children are born each year to permanent residents. The Behavior Screening Questionnaire (BSQ) of Richman and Graham is used as the method of detecting the presence of behavior problems in children. This questionnaire is included in a semi-structured parent interview schedule which includes demographic, child health, child temperament, family health, life events and social support variables. The children's behavioral and developmental status is assessed in a separate interview based on direct observation. Teacher ratings of the children's social adjustment are also obtained. The criterion by which the validity of the BSQ will be judged is clinical ratings made by a panel of child psychiatrists from all available data. A follow-up study of this cohort is planned to determine the incidence of behavior problems and the predictive validity of the screening questionnaire. The significance of this project relates to the primary and secondary prevention of childhood psychiatric disorder.